In order to determine more accurately the respective contribution to improvement of medication and counseling in treatment studies employing both modalities in outpatient treatment of cocaine abusers, this study aims at evaluating the influence of frequency of individual counseling on the effectiveness of cocaine abuse treatment. Subjects meeting DSM-III-R criteria for cocaine dependence will be randomly assigned (50 per group) to one of three treatment conditions plus urine monitoring for 12 weeks: a) twice weekly counseling, b) once weekly counseling, or c) counseling every other week. All counseling is delivered according to a specified therapy manual integrating aspects of interpersonal, cognitive, and behavioral approaches to drug abuse counseling. In the pilot phase, 27 subjects were treated to refine the manual and train counselors in the therapy. Treatment outcome will be assessed by urine toxicology twice weekly (collected under direct staff observation), breath alcohol analysis, self-reported drug use and cocaine craving, psychological symptoms, and psychosocial functioning (Addiction Severity Index). Follow-up at 3, 6, and 12 months will assess any longterm effects of treatment. To date, 107 subjects have entered the study, of whom 31 received initial evaluation only, with no actual treatment.